Pedal design - short support axle and spinning battery holder?

Former Member over 7 years ago

As I've been pondering some of the battery contact problems folks are having (me as well), it leads me to an interesting question. Is the problem just in the local electrical connections to the batteries? I've looked at how these are designed and it leaves me scratching my head a bit. I took the diagram that is provided by Garmin in their "How to rebuild the Vector 3 pedals" link and shifted around the components to try to understand the assembly relationships between the parts. I cut and pasted the photo of the axle into its correct location inside the pedal and the result seems striking. The final bearing surface is only about 15mm (or so) outboard of pedal center. That leaves a lot of cantilevered plastic (certainly composite filled plastic but plastic never-the-less) supporting the large pedaling forces onto the pedals. My pedals have developed "polished" spots in the stainless steel strike plates just under the outer and inner-most cross-point screws holding the plates to the tops of the pedals. This suggests a significant load applied well outboard of the outermost bearing support which could introduce flexing in the plastic pedal body.

Now, let's look at how power is delivered to the electronics of the pedal that are embedded inside the pedal axle. The end of the shaft has two small spring-loaded pins which contact a printed circuit board on the back of Part 4 which is screwed to the bearing retaining nut. These are not related to anything that you see when you're changing batteries; they're on the back side of the little printed circuit board that you see inside the pedal and which carries the center gold-plated battery fingers (negative battery terminal). It also has the two gold-plated pads that the outer legs of the battery carrier contact. If you stop and think about it, the pedal shaft is continuously turning relative to the pedal and the printed circuit board is tied to this shaft. That means that the entire battery carrier is spinning inside the pedal housing along with the shaft. The screw-in metal cap, however, is fixed to the pedals so the black plastic battery carrier needs to spin inside this metal cap! Seems okay at first but since the rider is also forcefully exerting against the pedals, if the unsupported plastic pedal body flexes as postulated above, this means that the battery carrier moves with respect to the center-of-rotation of the metal cap, potentially introducing all kinds of lateral forces. In a worst case, could the battery holder actually be forced into contact with the metal cap, causing drag and disrupting the battery contacts? You'll notice when you change the batteries that there is a lot of built-in mechanical "float" between the battery holder and the cap which would be needed to accommodate pedal deflections. However, if the center of rotation of the battery holder doesn't match that of the pedal during hard pedaling, there could be an inherent drag between the metal cap and the holder. There is a remarkably small amount of compliance in the positive battery contact arms which span the outermost surface of the outer battery (away from the bike) to the printed circuit board pads on Part 4. It seems this connection could be very susceptible to disruption.

Anyway, sorry for my rambling but it really struck me hard when I realized that the entire battery holder was fixed to the shaft, essentially causing it to spin relative to the plastic pedal housing and the metal end cap. You really don't think about this (or at least, I didn't) when replacing the batteries. This, combined by little support of the outboard portion of the pedals by the short pedal axle feels like a more significant design vulnerability than I had initially imagined. I'd be interested in your thoughts. ciq.forums.garmin.com/.../1338944.jpg

0 hansap over 7 years ago

There is always a Version 2.0 and I should have waited for it.

I bet that Vector 3 v2.0 will have a rechargeable battery inside the spindle. That would solve a lot of problems. It won't be easy to fit it in, because the electronics are already tiny, but I suppose that it can be done. Doing so could address the short spindle/long overhang concern that you have, because the charging connector won't take up as much space as the battery holder. The connection between the static spindle and the moving pedal body would be replaced by a connection between a static spindle and a static pedal body, because charging would take place while the bike stands around somewhere. While riding the source of power would move in sync with the printed circuit board.

Rechargeable battery in spindle, longer spindle, bearing further outboard, less flex under adverse conditions, no threads to kill, reliable power.

By the time v2.0 comes around the handling and the relative priorities of all the data streams [pedal to pedal - pedal to headunit ANT+ - pedal to headunit BLE and whatever else], and the buffering of each of those [there might be too much buffering], and all the other data related issues [smoothing, filling of gaps, ...] will be mostly sorted. v2.0 might still come with some faster hardware.

Faster processing, reliable data transmission.

0 Teriemer over 7 years ago

Interesting post CBDane. Could be you're right. Though most of the issues seems (to my knowledge) to be related to the right pedal. I would assume your view applies to both pedals.

But one thing we can agree on - which I have also stated numerous times - the design of V3 unfortunately is very poor. It is not robust and in particular the battery compartment is flimsy. No matter this is acknowledged, well-heard or not, this is a the massive issue for V3. And seeing all the issues now - I can only imagine where we will be in a year or 2 from here.

I simply don't get why Garmin doesn't come out of the bush, halting the production, redesigning the pedals, and clean up the mess. They would regain the lost faith from many customers. FWIW - I can report from a FB group where people are not recommended buying V3 - simply because the rumor is already out; they don't work. I don't know how many thousands are in that group, but probably a lot? And people are ditching V3 in favor of other pedals.

I just wish Garmin would communicate more. Fact is, it's been about 6 months since release and yet the pedals still have the same issues. Call it many or few - new people are still joining the forum with their issues every day. All my bells and alarm lights are flashing.

Garmin - please regain our trust. Tell us your plans. When can we expect a solidly working V3 (version 2)? With the competition constantly improving, time is running...

0 Mark2657 over 7 years ago

Anyway, sorry for my rambling but it really struck me hard when I realized that the entire battery holder was fixed to the shaft, essentially causing it to spin relative to the plastic pedal housing and the metal end cap.

Very interesting observation! So what keeps the pressure on the battery holder. Does the metal cap press on the spinning battery holder? If this is the case then there would be constant friction between the fixed cap and the battery holder.

0 Mark2657 over 7 years ago

The final bearing surface is only about 15mm (or so) outboard of pedal center. That leaves a lot of cantilevered plastic (certainly composite filled plastic but plastic never-the-less) supporting the large pedaling forces onto the pedals.

Your picture shows the end of spindle close to the middle of the pedal body. The 15 mm offset of the bearing may be an essential part of the design. The strain gauges are mounted inside the spindle, so in order to enhance the flex the spindle, it would need to be shorter than the pedal body. The pressure from the pedal needs to be applied close to the end of the spindle in order to enhance the bending action.

0 Former Member over 7 years ago

@<Mark2657, Yes the metal cap presses on the spinning battery holder. You can't see it but there seems to be a single point of contact to minimize friction (like the spinny tops I collect on my desk at work). A potential issue, however, is if this point of contact is driven slightly off center (momentarily) by pedal flex, it seems like this could introduce cyclic side loads on the spring-loaded battery carrier. I think the amount of fretting shown in the Garmin advisory is pretty impressive and I don't see the source for that kind of vibration unless the battery carrier was getting jostled on each pedal revolution. You could be well right about the length of the axle and the strain gauge design.

UPDATE: I posted the above thoughts into the construction geometry after I had experienced two bad commutes with many drop-outs on my right pedal. After this investigation, I reassembled my pedal, carefully applying a very (very) small amount of pure silicone grease (Corning vacuum compound) to the two spring loaded pins behind the circuit board. All of the gold pads looked a little questionable so I carefully polished them (front and back of the PCB) with a small Pentel pencil eraser (watch for rubber debris from this process). It almost seemed like there was a light film of material over the contacts which I wonder could have outgassed from the adhesive used to cement the PCB to the plastic carrier that screws to the bearing retaining nut. I then put light coats of the same grease between the batteries and each electrical contact before reassembling. I'm quite happy to report that I had two perfect rides (home and back to work today) with zero issues in the data.

My history with the Vectors is a set of 3s (one-sided) with 200 miles of not one drop-out or other problem. I upgraded to the 3 last week and began to experience these issues right away. Not that it matters, but my job in my spare time is designing electronics (circuit design, PCB design, and embedded programming) for a motorsports equipment manufacturer.

0 Mark2657 over 7 years ago

@<Mark2657, Yes the metal cap presses on the spinning battery holder. You can't see it but there seems to be a single point of contact to minimize friction (like the spinny tops I collect on my desk at work).

Now I understand why T_Smit did not like the idea of putting cork under the contact spring. This would put more pressure on the interface between the single point battery holder contact and the cover, which could greatly accelerate wear. The reason the cork works is probably due to the mechanical damping it introduces into the battery holder. You are right that the fretting seen on the batteries indicates that they are really moving around quite a bit.

I think you are on to something with your out gassing observation. I had a similar issue with an Amazon Kindle which had an integrated battery/cover. It was shipped in a sealed plastic shrink wrap bag and you could smell the out gassing from the plastic as you opened it. The battery/cover connected to the Kindle with gold spring contacts, just like the Vector 3, and initially would not work. I used an eraser to clean the contacts of out gassing residue and it then worked perfectly. The Vector 3 contacts could be contaminated by the out gassing of the glue or it could be the out gassing of the plastic from the pedal body itself. I think that this is a very important finding. It could explain why some people have issues that never seem to go away while others see no issues at all. Cleaning the out gassing residue may be the solution.

0 Former Member over 7 years ago

Now I understand why T_Smit did not like the idea of putting cork under the contact spring. This would put more pressure on the interface between the single point battery holder contact and the cover, which could greatly accelerate wear. The reason the cork works is probably due to the mechanical damping it introduces into the battery holder. You are right that the fretting seen on the batteries indicates that they are really moving around quite a bit.

The other issue that I see with the cork idea is that it actually works against the two positive battery contact points (at the end of the small bars under the insulating tape) between the battery carrier and the small PCB, trying to hold them further off from the board. These are even less compliant than the gold fingers that make up with the negative terminal of the battery stack.

I think you are on to something with your out gassing observation. I had a similar issue with an Amazon Kindle which had an integrated battery/cover. It was shipped in a sealed plastic shrink wrap bag and you could smell the out gassing from the plastic as you opened it. The battery/cover connected to the Kindle with gold spring contacts, just like the Vector 3, and initially would not work. I used an eraser to clean the contacts of out gassing residue and it then worked perfectly. The Vector 3 contacts could be contaminated by the out gassing of the glue or it could be the out gassing of the plastic from the pedal body itself. I think that this is a very important finding. It could explain why some people have issues that never seem to go away while others see no issues at all. Cleaning the out gassing residue may be the solution.

There is quite a bit of glue holding the PCB to the plastic carrier that screws to the the bearing retaining nut. If you peer down into the pedal body, you'll see that it is squeezed out into a small bead along each side of the circuit board.

BTW: Another two commute legs with zero dropouts of pedals or cycling dynamics (total of four since the clean-up). Feeling pretty happy for now!

0 Mark2657 over 7 years ago

I think that you may have solved the mystery of the battery holder issues. The out gassing on the contacts and PCB traces would explain why the problem is not solved by replacing batteries, bending contacts, putting oil and other tricks.

If you have the time, I would suggest that you start a new thread proposing your idea as a solution to the battery holder/dropout issues. If you could take some step by step pictures on how to clean the contacts, as well as the recommended materials to use, I think that it would be very much appreciated by everyone.

0 Mark2657 over 7 years ago

BTW: Another two commute legs with zero dropouts of pedals or cycling dynamics (total of four since the clean-up). Feeling pretty happy for now!

Are your pedals still functioning well after cleaning the contacts?

0 Former Member over 7 years ago

Are your pedals still functioning well after cleaning the contacts?

Not one drop-out (or even a hiccup) since. The data has been so smooth, complete, and well-behaved that I decided to do an experiment yesterday evening. For the last mile home, I intentionally tried to place most pedaling force on the left pedal to make sure that I saw an effect. Basically, I caused the balance data to scatter all over the map (it's very hard to pedal with one leg efficiently) with a bias toward the left pedal. Working as hoped and expected!